File size: 3,753 Bytes
35bb6f4 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 | from __future__ import annotations
from typing import Literal
from pydantic import BaseModel, Field
# --- OpenAI-compatible Request ---
class OpenAISpeechRequest(BaseModel):
model: str = "neutts-nano-q4-gguf"
input: str = Field(..., min_length=1, max_length=10000)
voice: str = "jo"
response_format: Literal["mp3", "opus", "aac", "flac", "wav", "pcm"] = "mp3"
speed: float = Field(default=1.0, ge=0.25, le=4.0)
stream: bool = False
# --- OpenAI-compatible Responses ---
class VoiceInfo(BaseModel):
voice_id: str
name: str
language: str
gender: str
custom: bool = False
available: bool = True
preview_url: str | None = None
class VoiceListResponse(BaseModel):
voices: list[VoiceInfo]
class ModelInfo(BaseModel):
id: str
object: str = "model"
created: int = 0
owned_by: str = "neuphonic"
language: str | None = None
backend: str | None = None
supports_streaming: bool = False
backbone_device: str | None = None
codec_device: str | None = None
class ModelListResponse(BaseModel):
object: str = "list"
data: list[ModelInfo]
class ModelDetailResponse(ModelInfo):
loaded: bool = False
codec: str | None = None
# --- Model Management ---
class LoadModelRequest(BaseModel):
model_id: str
codec: str | None = None
backbone_device: str | None = None
codec_device: str | None = None
class UnloadModelResponse(BaseModel):
model_id: str
status: str = "unloaded"
class ModelLoadTaskResponse(BaseModel):
task_id: str
model_id: str
status: str
progress_message: str = ""
error_message: str = ""
elapsed_seconds: float = 0.0
class SwitchDeviceRequest(BaseModel):
backbone_device: str | None = None
codec_device: str | None = None
class LoadedModelInfo(BaseModel):
model_id: str
codec: str
backbone_device: str
codec_device: str
language: str | None = None
backend: str | None = None
supports_streaming: bool = False
class LoadedModelsResponse(BaseModel):
models: list[LoadedModelInfo]
class RegistryModelInfo(BaseModel):
model_id: str
repo: str
language: str
backend: str
supports_streaming: bool
description: str
loaded: bool = False
class ModelRegistryResponse(BaseModel):
backbones: list[RegistryModelInfo]
codecs: list[dict]
# --- Voice Management ---
class VoiceEncodeRequest(BaseModel):
codec: str | None = None
class VoiceUploadResponse(BaseModel):
voice_id: str
status: str = "uploaded"
message: str = ""
language: str = "unknown"
gender: str = "unknown"
class VoiceDeleteResponse(BaseModel):
voice_id: str
status: str = "deleted"
class VoiceEncodeResponse(BaseModel):
voice_id: str
codec: str
status: str = "encoded"
# --- Health & Debug ---
class HealthResponse(BaseModel):
status: str = "ok"
version: str = "0.1.0"
models_loaded: int = 0
class SystemDebugResponse(BaseModel):
cpu_count: int
cpu_percent: float
memory_total_gb: float
memory_used_gb: float
memory_percent: float
gpu_available: bool
gpu_info: list[dict] | None = None
torch_version: str | None = None
cuda_version: str | None = None
cuda_driver_version: str | None = None
gpu_detected_but_unusable: bool = False
gpu_fix_instructions: str | None = None
models_loaded: list[str]
voices_available: int
# --- Error ---
class ErrorResponse(BaseModel):
error: dict
def make_error(message: str, error_type: str = "invalid_request_error", code: int = 400) -> dict:
return {
"error": {
"message": message,
"type": error_type,
"code": code,
}
}
|